Latin American feminists brought up the issue of violence in the 1970s under military rule or situations of armed conflict. These contexts made feminists specifically concerned with state violence against women. Women's organizations pointed to torture and rape of political prisoners and the use of rape as a weapon of war and connected these forms of violence to deeper societal patterns of subordination and violence against women in both the private and public spheres. Processes of democratization in the region brought new opportunities to institutionalize norms to end violence against women (VAW), and in many countries feminists managed to get the issue on the political agenda. In the mid 1990s, the region pioneered international legislation on VAW that uniquely included state-sponsored violence. The Inter-American Convention on the Prevention, Punishment and Eradication of Violence against Women (1994) established an international obligation for states to prevent, investigate, and punish VAW regardless of whether it takes place in the home, the community, or in the public sphere. While Latin American governments massively ratified this convention, national legislation was not brought in line with the broad scope of the international convention. This points to the complex and often contradictory dynamics of institutionalizing norms to oppose VAW in multilevel settings.
Much of the literature that deals with international norm development depicts norm diffusion as a rather mechanical process. The central concern is how norms emerge on the international level and how these norms then become adopted and implemented at the national level. More recently, a number of authors argue that we need to develop more complex approaches to transnational norms diffusion processes (Krook and True 2012; Montoya 2013; Zwingel 2012). In this article I attempt to contribute to a more adequate theoretical understanding of norm diffusion building on the case of feminist norm setting on violence against women under conditions of multilevel governance in Latin America. I argue that, in particular, three central assumptions prevalent in the global norm diffusion literature are problematic to understand these complex multilevel processes of norm institutionalization.
First, in the vast body of literature that emerged on the way norms diffuse, these processes are often understood as linear global to local processes (Zwingel 2012). Such a model tends to treat national and international policy arenas as separate and hierarchical. I want to argue that instead we need a model that integrates multiple and often parallel multilevel efforts to institutionalize norms. The different levels that feminists may target in their efforts to end VAW are coconstituted and have to be studied as such. This brings me to the second point: the top-down transmitter-receptor model fails to take into account that the origins of norms are often multiple, and therefore norm diffusion may be multidirectional. Norm diffusion processes can be bottom-up, top-down, or horizontal. A third problematic point is that the transmission model pays little attention to how norms are subject to interpretation, contestation, and negotiation and therefore change as they diffuse.
More research is needed to refine and reconceptualize the complex transnational-domestic processes in which norms emerge in multiple interrelated locations. In this paper, I seek to contribute to a more dynamic understanding of norm diffusion by examining the processes of multilevel norm setting to end VAW in Latin America. I propose that norm-setting processes that oppose VAW in Latin America are particularly suited and interesting cases to challenge the prevalent mechanical assumptions about international norm development. Latin American processes eliminating violence against women preceded many other transnational processes by recognizing state responsibility. Struggles to make the state accountable for violence in the private or domestic sphere met with resistances. The experience of military regimes that had invaded private life created a strong noninterventionist stance among many political actors. Yet, despite these difficulties the region adopted the first binding international legal instrument that specifically addresses VAW. The Inter-American Convention on the Prevention, Punishment and Eradication of Violence against Women (1994) uniquely included state-sponsored violence. To untangle the complex transnational-domestic processes that happen both prior to and after the formulation of international legislation I look at two Latin American countries: Brazil and Uruguay. From a top-down norm diffusion approach that looks at state behavior after the establishment of an international norm these countries could be depicted as, respectively, a leader or early adopter (Uruguay) and a noncompliant or reluctant state (Brazil).
Yet, by taking the Inter-American Convention as the starting point for analyzing norm-setting processes we ignore the role of Brazil as a regional pioneer prior to the adoption of the OAS Convention. In addition, we fail to see that Uruguay, instead of a willing adopter, has been notoriously slow and limited in its response to VAW.
Before I turn to these empirical cases, I first discuss theoretical approaches to norm establishment and diffusion and delve deeper into some central problematic assumptions in the literature on international norm diffusion. Next, I argue against a static conceptualization of norm-diffusion processes and propose an alternative approach to norm diffusion as multilevel and multidirectional in which norms are subject to ongoing contestation and change. I then turn to my empirical cases starting with early Latin American feminist efforts to institutionalize norms to end violence both at the national and regional level. I show how the Inter-American Commission on Women as an autonomous body within the OAS responded to these processes of national and transnational mobilization and started to explore its own potential role. I argue that the initiatives of Inter-American Commission on Women were embedded in and overlapped with other regional and national processes that help explain how the Commission succeeded in establishing groundbreaking international legislation within a relatively short time frame. I then examine the problematic national trajectories in Brazil and Uruguay after the adoption of the Inter-American Convention. In both countries measures were taken that drastically narrowed down and distorted the progressive framing of the Convention. It took feminists in both countries almost a decade to pressure the state to adopt legislation that is more in line with the Convention, yet to date neither Brazil nor Uruguay has fully adopted the Convention. This indicates that processes of norm setting in multilevel settings do not resemble linear and mechanistic models of norm diffusion. Instead, we need to pay attention to contestation and negotiation of norms. I then conclude with a discussion of implications for the literature on norm diffusion.
THEORETICAL DISCUSSION: UNDERSTANDING MULTILEVEL NORM DYNAMICS
A vast body of International Relations scholarship has developed that explores the creation of international norms and their dissemination to the domestic realm. In this global norm diffusion literature an international perspective prevails that mainly focuses on the impact of international institutions on national policies. Finnemore and Sikkink's prominent life-cycle model (1998) depicts norms as emerging, cascading, and then becoming internalized. In this process, states “are socialized to accept new norms, values and perceptions of interest by international organizations” (Finnemore and Sikkink 1998, 5). This overly optimistic and rather mechanical model obscures the fact that even if certain norms are formally defined and adopted in international legislation they often have very uneven effects (Zwingel 2012). International legislation may indeed be effectively used to push for or improve national legislation but may also have reverse effects and lead to a stagnation of existing domestic policy processes or a reversal of previous gains. Domestic dynamics are largely ignored. In this section, I go deeper into some of the problematic assumptions that prevail in norm diffusion literature.
First, diffusion is often modeled as a global to domestic linear model in which these levels are treated as separate and hierarchical (Zwingel 2012). The norm-cycle model, despite the cycle metaphor, principally views diffusion as a “one way process in which norms emerge and are then communicated and internalized” (Krook and True 2012, 108). Several authors have come up with more dynamic models. Keck and Sikkink (1999) developed the often-cited “boomerang pattern,” in which societal groups that face unresponsive states seek the support of international organizations, which in turn place pressure on reluctant states to change their practices. More recently, Montoya (2013) developed a theoretical framework that distinguishes four different pathways. The first she labels bottom-up domestically driven reform, which is initiated by strong societal mobilization that successfully pressures the government to respond. The second pattern is top-down domestically driven reform by strong advocates within state policy institutions. A third model is transnationally driven reform that requires a form of interaction between domestic actors and international advocates. A fourth pathway is internationally driven reform, which “arises when the results of transnational mobilization of a few countries garners an international response that spills over into countries lacking active or effective domestic advocacy” (Montoya 2013, 36). This theoretical model acknowledges that interactions between international and domestic actors may take multiple forms. Yet, it still treats the global and local level as separate and hierarchical instead of overlapping and interrelated. Also, by defining reform as the end product of each pathway, it does not recognize diffusion processes as ongoing struggles in which reform triggers new processes of contestation or struggle.
Second, much of the norm diffusion literature is concerned with how international treaties and legislation, once established, affect national contexts. Norm diffusion is then conceptualized as a transmission model in which international organizations are the sources and national states figure as receivers. This not only underestimates the domestic creation of norms (Zwingel 2012), but also fails to take into account that norms may have multiple origins and multiple receivers. Because norm creation is dispersed and takes place at different levels and in different arenas, the directions in which norms diffuse therefore are also multiple. Ideas and models may diffuse bottom up, top down, or through horizontal channels. Instead of taking international norms as a starting point, the important empirical question is where different ideas come up and how they are taken to different spaces and places.
Third, the transmission model tends to treat norms as clear-cut and fixed packages that are transplanted in whole cloth from one site to another (Givan, Roberts, and Soule 2010). The metaphoric language used in many of the approaches to norm diffusion views norms as static entities that are advocated (Keck and Sikkink 1998; True and Mintrom 2001), adopted, cascaded, and internalized (Finnemore and Sikkink 1998; Risse, Ropp, and Sikkink 1999). Also, norms appear to have a life of their own, or even agency (Bucher 2014). Norms have a “life cycle,” travel, or exert influence. There is a growing recognition that norms are not fixed but subject to change. As Krook and True (2012, 110) argue, “the acceptance of a norm may initiate rather than resolve struggles over its exact content,” leaving space for actors to adapt, interpret, and challenge the meaning of norms. Norms, thus, should be seen as processes in themselves and, as such, part of the multilevel process instead of fixed end products that are transferred to a different level. These discursive norm dynamics should be theorized as central to norm diffusion to help us better capture the complex and often contradictory processes that are inherent to multilevel norm setting. A model that seeks to study norms as discursive dynamics requires a “longer-term perspective” (Krook and True 2012) that not only takes into account the processes of norm diffusion after the institution of an international norm, but also norm-setting processes that precede the establishment international legislation. In the rest of this paper, I therefore look at almost four decades of struggle to institutionalize norms to end violence in Latin America. These long-term processes make clear that norm diffusion processes are multidirectional and that ongoing contestation is the key dynamic of norm diffusion.
EARLY FEMINIST NORM SETTING: FACING A PRECARIOUS STATE AND A FAVORABLE INTERNATIONAL CONTEXTS
In this section I examine early Latin American efforts to institutionalize norms that opposed violence against women at the national and transnational level that led up to the establishment of the Inter-American Convention. These national and regional processes coevolved and are interrelated. Feminist efforts to engender new spaces at the national level were influenced by parallel processes at the international level (UN) and regional (Latin American) level. The 1975 International Women's Year Conference in Mexico City and the UN Decade of Women that followed (1976–1985) gave an important impetus to feminist organizing at both the national and regional level. The preparations for this conference revitalized existing organizations, established new groups, and made gender a political issue (Jaquette 1994, 4). In 1982, the first Latin American and Caribbean feminist meeting (Encuentro) was held in Bogotá. Participants discussed the issue of violence and launched joint campaigns and research. These Encuentros also helped diffuse experiences, national campaigns, and innovative models for service provision from one context to another. In addition, regional networks became an important vehicle for regional cooperation. In 1987, the regional rights organization CLADEM (Latin American and Caribbean Committee for the Defense of Women's Rights) did a regional evaluation of national legislation in relation to VAW (Guerrero 2002, 38). While CLADEM started as a network for the exchange of information and construction of a joint database on the legal situation of women, it became increasingly involved in advocacy and lobbying both at the national and regional level and made an important contribution in understanding VAW as a human rights issue (Chiarotti 2006). In 1990, during the fifth regional Encuentro, the Latin American and Caribbean Network against Domestic Violence and Sexual Violence was founded, linking 379 national organizations. Regional networking was central to the development of a shared framing of the problem as activists began to analyze the regional pattern of military dictatorship and authoritarianism in the public sphere and its relation to the prevalence of authoritarian and patriarchal relations in society and in the family. From this, Latin American feminists began to develop a broad framing of the problem of violence that included physical, sexual, and psychological harm to women but also more indirect acts of humiliation and discrimination keeping women in subordinate roles. Violence, they argued, not only constituted a violation of their human rights, but also a major obstacle for women to be free and equal citizens in the newly developing democracies (Jaquette 1994; Keck and Sikkink 1998).
This pattern of transnational mobilization, embedded in national processes, eventually paved the road to international legislation. Latin American feminists played a very active and pioneering role in global conversations on VAW and became key players in spreading information on the issue (Keck and Sikkink 1998, 170, 179) Yet, while transnational processes helped to build a strong regional movement with a collective agenda and a shared understanding of the problem of VAW, national efforts to get VAW on the political agenda were met with uneven responses, as the cases of Brazil and Uruguay illustrate.
Brazilian feminists started to campaign the issue of violence in the late 1970s. Several publicly visible incidences of “crimes of passion” committed by male celebrities and the murders of women by jealous husbands spurred protest and outrage in several Brazilian cities (Alvarez in Jacquette 1994, 31). Starting from the issue of femicide, Brazilian feminists started to extend their focus to psychological, physical, and sexual partner violence. In October of 1980, a feminist organization in São Paulo opened the SOS Mulher office; this office was the first entity in Brazil that tended to abused women. SOS Mulher offered legal and psychological counseling and organized information and awareness campaigns. The assistance model of SOS Mulher rapidly spread across Brazil and traveled to neighboring countries like Uruguay and Argentina. The experiences of the SOS Mulher groups, mainly attended by women that suffered from partner and family violence, helped influence their understanding of the problem of violence against women. The group framed violence against women as resulting from a patriarchal structure of domination and pointed to the cultural bases underpinning the subordinate position of women (Pitanguy 2002, 809). This interpretation clashed with the discourse prevalent in the police force and judicial system. SOS Mulher pointed out that victims of abuse did not receive adequate treatment from the police and started to gather evidence of the difficulties that women encountered in reporting to the police. This research became a driving force behind the creation of special police units (Santos 2004). In 1985, SOS Mulher helped to establish the first all-women police unit specialized to assist women who had suffered from abuse: the Delegacia Especializada de Atendimiento a Mulher (DEAM) in Sao Paulo (Nelson 1996). This example rapidly spread across Brazil (in five years the number grew to more than 200 police units) and was adopted in other parts of the region (Colombia, Ecuador, Uruguay).
The early initiatives to get violence against women on the public and political agenda developed within a favorable political context. As the opposition against the authoritarian regime increased and gained electoral strength in the mid-1970s and gained control of some state and municipal governments after 1982, feminists started to engage more with institutional politics (Alvarez 1994). Opposition parties made gestures to create alliances with women's organizations and adopted some feminist issues. Women's divisions were established within most opposition parties. While some activists insisted on organizational and ideological autonomy from parties and the state, many other feminists embraced these new opportunities. Women within the opposition party PMDB (Partido do Movimento Democratico Brasileiro) made a proposal for a Council on Women's Condition that would “serve as an instrument for a global policy destined to eliminate the discriminations suffered by W = women” (Alvarez 1994, 37).
After the victory of the PMDB in Sao Paulo, their proposal was adopted in 1983. The new council was given extensive advisory powers but very limited executive powers and no independent budget. The council's political room for maneuvering thus depended on the goodwill of the state governor. Yet, many women's organizations viewed the council as an opportunity to influence the political agenda and started to channel their demands to the council which created a Commission on VAW that then successfully lobbied for the creation of the earlier-mentioned special police units (DEAMs). In 1985, when the first civilian president in 21 years took office, a similar council was instituted at the national level: the National Council on Women's Rights. This council provided the women's movement with new points of access to the state and also financial support for their projects. During the first years of its creation, the council was successful in setting the policy agenda and was particularly active on the issue of VAW. It launched a national awareness campaign and compiled a national dossier on the impunity of acts of VAW (Alvarez 1990). The council was staffed by feminists and held close contact with women's organizations. During the process of formulating a new constitution, the CNDM acted as “an effective de facto women's lobby” and organized conferences and public forums all over Brazil where proposals for the constitution were formulated and culminated in a letter to the constituent assembly (Waylen 2006, 1212). The demands expressed in this letter were supported by a newly created women's caucus in the Chamber of Deputies. This alliance succeeded in incorporating many women's demands in the final constitution,
including, almost uniquely in Latin America, a provision on family violence (Waylen 2006, 1212). According to the 1988 constitution, “the State shall ensure assistance to the family in the person of each of its members, creating mechanisms to suppress violence within the family” (1988 Article 226, par 8).
While this made family violence a state responsibility, it also limited the problem of violence against women to the realm of the family and did not pay attention to the gender dimension of the problem.
After this initial “reformist euphoria,” progress stagnated in the late 1980s, early 1990s (Alvarez 1990). Feminist attorneys denounced the failure to implement the reforms of the civil and penal codes envisioned by the 1988 constitution “leaving Brazilian women in the precarious state of having constitutional rights which are not reflected in the codes designed to enforce those rights” (Alvarez 1994, 48). The penal code, for instance, still defined rape as a crime against custom rather than a crime against an individual victim. Also, prosecution and conviction rates remained low. The advances in institutionalizing norms on VAW and the new institutional women's spaces such as the councils on the status of women were gradually dismantled (Alvarez 1994).
The issue of violence against women appeared on the agenda of Uruguayan women's organizations almost a decade later than in Brazil and was inspired by the example of Brazil and other examples in neighboring countries. Yet, compared to Brazil, mobilization was less strong, and women's organizations were very cautious to bring up the issue. As one of the initiators of the Uruguayan SOS Mujer explained, the resistance in society to feminism made them chose a careful strategy and avoid the use of the word feminist (Repetto 1992). It also proved particularly problematic to make the state accountable for violence in the private or domestic sphere. The experience of state intrusion of all areas of private life during military rule made opposition parties hesitant to include the issue of domestic violence since they feared to be accused of intervening in the domestic sphere (Johnson 2002, 109).
In contrast to the Brazilian case, the phase of transition created few openings for women's organizations to get the issue of VAW on the political agenda. The Frente Amplio, a coalition of left-wing parties, did not want to take up the issue. as they argued that they did not want to be accused of intervening in the domestic sphere or privacy of the home (Johnson 2002). Women struggled to get the issue on the agenda of the Concertación Nacional Programática (CONAPRO), a joint effort of political parties and social organizations to reach a national consensus on how to establish a new democracy after the period of military dictatorship (Corbo 2007). The CONAPRO committee on the Status of Women, in which both women from political parties and women's organizations participated, recommended to investigate the incidences of domestic violence and to launch an awareness and prevention campaign. While the CONAPRO executive body approved these recommendations, the Colorado Party government, which took office after the elections in 1985, did not develop initiatives in relation to violence against women. In the face of this political nonresponse, women's organizations started to focus on autonomous service provision. In 1988, a first organization specifically dedicated to the issue of violence, SOS Mujer, following the example of Brazilian pioneer SOS Mulher, started a support service including legal and medical care and treatment both to victims and perpetrators. Additionally, women's organizations jointly organized demonstrations and public campaigns on the International Day against Violence (November 25) to raise consciousness and denounce the lack of government response. The Consejo Nacional de Mujeres (CONAMU) a national organization of liberal feminists, travelled to Brazil to study the functioning of the recently founded special police units for women. When the government established the National Women's Institute (Instituto Nacional de Mujeres) in 1987, CONAMU successfully lobbied for the adoption of this Brazilian example.
Yet in 1990 the right-wing National Party formed a new government that abolished the Women's Institute and state financing for women's initiatives. Two years later the government established a new Institute for Family and Women with a weak mandate and a very limited budget (Johnson 2002). Additionally, existing precarious links with women's organizations were discontinued. While the new institute took up the issue of violence, it only focused on family violence and degendered the issue. Inspired by the successful regional example of feminist networking, Uruguayan activists decided to turn to “issue-based forms of coordination” to mount political pressure (Johnson 2002). In 1992, eight women's organizations working on the issue of violence against women formed a national network against domestic violence (Red Uruguaya contra la Violencia Domestica y Sexual, RUDVS). This group held close links to the recently created Latin American and Caribbean Network against Domestic Violence and Sexual Violence. At the same time, a group of feminist academics founded Espacio Feminista, a special working group on violence against women that combined research with lobbying efforts (Johnson 2002). Both groups made efforts to get violence against women on the political agenda by mobilizing the emerging transnational human rights framing of the problem. In a letter to the Human Rights Committee of the Chamber of Deputies (April 1993), Espacio Feminista wrote: “Our aim is that violence against women be understood as a global political problem. We consider this violence to be systematic in nature, which is what forces us to see it as a question of human rights” (cited in Johnson 2002, 108). The right-wing government and its neoliberal orientation remained unresponsive to feminist demands for state intervention and service provision. Espacio Feminista then turned to the innovative strategy of using international litigation. In September 1993, the group filed a complaint against the Uruguayan government at the Inter-American Commission of Human Rights, under the provision of the Inter-American Human Rights Convention (Ewing 1994; Johnson 2002, 111). This was the first claim made against a government for failing to protect the human rights of individuals in cases of domestic violence before the OAS Convention on Violence against Women created this possibility. Although the case was not won, it had important regional repercussions, as it opened the possibility of using international conventions as instruments to make the state responsible to end violence against women.
This reconstruction of early national and regional dynamics makes clear how national and regional processes coevolved. Feminist efforts to set norms to end violence occurred simultaneously at different interrelated levels. Ideas and models to address violence developed in different local situations, crossed national boundaries, and emerged from encounters in new transnational spaces. This coconstruction of knowledge makes it difficult to identify a single source or transmitter of ideas and models. While Brazil had been pioneering innovative models to address VAW that diffused cross-nationally, Uruguayan feminists were the first in the region to use the Inter-American Human Rights Convention as an instrument to make the state accountable for domestic violence. Specific experiences of state violence, processes of democratization, and cultural tolerance toward male violence influenced the way activists in Brazil and Uruguay understood the problem of violence. Yet, these ideas and models did not emerge in closed national contexts but in interrelated and nested local, national, and global spaces. The UN Decade for Women created new opportunities, which were vital for local women's organizations that experienced difficulties in dealing with precarious and sometimes reluctant national states. Also, international frameworks provided legal and conceptual tools to frame violence against women as a human rights issue. Yet, these ideas and efforts to end VAW and make the state accountable were not equally successful in the two countries. While for Brazilian feminists the processes of democratization opened important possibilities to get the issue of VAW on the political agenda, Uruguayan women's organizations were less successful in engendering the transition process. This resulted in different state-movement relations and different attitudes toward the state as a potential ally in the struggle against violence.
The Establishment of International Legislation to End VAW
The establishment of international legislation in 1994 does not mark a starting point as the previous section made clear, nor can be seen as an end product. In this section I argue that the OAS Convention should be seen as a response to the national and regional efforts to put the issue of VAW on the political agenda. Once established, it generated new norm dynamics as I will argue later.
The Inter-American Commission of Women of the Organization of American States (OAS) took up the issue of VAW in 1988 (Meyer 1999). The Inter-American Commission (known by its Spanish acronym CIM, Comisión Interamericana de Mujeres) is an autonomous organization within the OAS that is made up of 33 delegates who are designated by the governments of the OAS member states.
The CIM meets every two years and provides technical and advisory support to the OAS member states on the implementation of key conventions and other agreements regarding women's rights in the region. Despite its political nature, limited mandate, and small staff, CIM has been able to set its own agenda and draft a number of Inter-American Conventions on women's rights (Friedman 2009).
The Inter-American Convention of VAW was established within a period of six years (Friedman 2009; Meyer 1999). This rapid success is remarkable in view of CIM's limited mandate and the budget and staff cutbacks that CIM faced in the 1980s (Meyer 1999; Poole 2013). I argue that several factors explain the rapid establishment of this new international norm.
First, CIM was able to build upon the important transnational and domestic work of women's organizations in the region to make the problem of VAW visible. National experiences provided CIM with important insights in institutional possibilities and obstacles. National and regional expertise helped to design the language and structure of the Convention.
Second, the center of inter-American action has been the work of the human rights mechanisms. The 1948 American Declaration of the Rights and Duties of Man and the 1969 American Convention on Human Rights established principles against state impunity in cases of human rights violation under authoritarian regimes and armed conflict. The Inter-American Commission on Human Rights (IACHR) and the Inter-American Court of Human Rights played an important regional role in denouncing state violence against civilians. This track record and status provided CIM with important enforcement mechanisms.
Third, the development of the inter-American norm cannot be understood without taking into account the wider international framework, in particular that of the UN. It was within this context that the president of CIM, Milagro Azcúnaga de Meléndez (El Salvador), during a 1989 meeting of the Convention on the Elimination of all Forms of Discrimination against Women (CEDAW) announced that the lack of attention to the issue of VAW in CEDAW was the reason that CIM started a regional campaign to establish international legislation (Poole 2013). CIM's campaign then overlapped with a key international development on violence against women: the organizing around the 1993 UN Conference on Human Rights in which violence against women became recognized as a human rights violation (Friedman 2009). This human rights framework is also the basis of the Inter-American Convention. When the UN General Assembly adopted the Declaration on the Elimination of Violence against Women (A/RES/48/104) in December 1993, drafting for the Convention of Belém do Pará was nearing completion (Friedman 2009).
Meyer argues that the process to establish the Inter-American Convention on VAW builds upon CIM's “traditional strategy” (Meyer 1999, 66) of pressing governments to bring national laws in compliance with new international norms.
It also clearly aligned with the work of regional and national women's organizations that had been active on the issue. To prepare the convention, CIM engaged in an open and transparent process that included the consultation of a wide range of civil society organizations (Poole 2013).
The Convention adopted the broad definition that came out of this consultation that defines violence against women as “a violation of their human rights” (Meyer 1999). This resulted in a “pathbreaking international human rights norm” (Meyer 1999, 69) that makes the state responsible for defending and protecting rights of women. According to its preamble, violence against women should be understood as a “manifestation of the historically unequal power relations between women and men.” The convention adopts a broad and inclusive definition of violence against women as “any act or conduct, based on gender, which causes death or physical, sexual or psychological harm or suffering to women, whether in the public or the private sphere.” This was narrower than the definition proposed by regional women's organizations that also contained “indirect acts that intimidated or humiliated women, maintained them in sex-stereotypes roles, or denied them human dignity, whether or not these acts caused physical or mental injuries” (Keck and Sikkink 1998, 172). Violence may occur in the private sphere, defined as “the family or domestic unit or within any other interpersonal relationship, whether or not the perpetrator shares or has shared the same residence with the woman,” in the sphere of the community “including, among others, rape, sexual abuse, torture, trafficking in persons, forced prostitution, kidnapping and sexual harassment in the workplace, as well as in educational institutions, health facilities or any other place,” and in the sphere of the state referring to all violence “that is perpetrated or condoned by the state or its agents regardless of where it occurs.”
Chapter II of the Convention states that women have the right to be free from violence, which includes the right be free from all forms of discrimination and “to be valued and educated free of stereotyped patterns of behavior and social and cultural practices based on concepts of inferiority or subordination.” State parties “recognize that violence against women prevents and nullifies the exercise of their human rights” (Convention, II, 5). The Convention requires the state to investigate, sanction, and prevent all forms of VAW. Chapter III of the Convention clearly outlines the duties of states to “pursue, by all appropriate means and without delay, policies to prevent, punish and eradicate such violence.” Additionally, states must ensure “effective access to restitution, reparations, or other just and effective remedies.” Governments should also modify legal or customary practices that sustain the persistence and tolerance of violence against women; develop programs to raise awareness; train justice, police, and law enforcement officers; and provide appropriate specialized services for women. States are obliged to send national reports the Inter-American Commission of Women explaining how national policies are brought in line with the Convention. Also, individuals or organizations may lodge petitions with the Inter-American Commission on Human Rights for alleged violations of states' responsibilities under the Convention (Article 7).
The Convention was adopted by acclamation during the General Assembly in Belem do Pará on June 9, 1994. It was rapidly ratified by most OAS members.
Between 1994 and 2000, most Latin American states ratified the Convention, and many countries passed national legislation. These national laws did however have a limited focus on the domestic realm and family violence and did not address violence in the sphere of the community or the state (Larraín 1999). Also, the gender-dimension of the problem was systematically ignored. In the next section, I examine the complex and ongoing process of adaption and contestation of the OAS Convention in Brazil and Uruguay. This demonstrates that prevailing linear and mechanical top-down diffusion models fall short of understanding these complex norm dynamics.
Postratification Processes in Brazil and Uruguay
The 20-year period after the adoption of the Inter-American Convention can be divided in two periods because in 2004 the OAS General Assembly adopted an additional monitoring mechanism to put more pressure on states to implement the Convention. This “Mechanism to Follow-up on Implementation of the Inter-American Convention on the Prevention, Punishment and Eradication of Violence against Women” (MESECVI) is a continuous and independent evaluation process in which states have to report to a committee of independent experts who then review the progress of each state. The committee has published two hemispheric evaluation reports (in 2008 and 2012) that point to gradual progress, in particular on the issue of domestic violence. Also, some countries in the region developed comprehensive laws on VAW to facilitate unified and consistent treatment of the different forms of violence against women in public policy and justice.
Yet, many states still fail to address violence in the sphere of the community and the state. Also, the committee of experts expressed their concerns regarding the political will of the states to support the mechanism.
A closer look at the cases of Brazil and Uruguay reveals the ambiguity and reluctance of governments to adopt the OAS Convention. While the governments of both counties ratified the Convention, they also took measures that drastically narrowed down and distorted the progressive and broad framing of the Convention. It took feminists in the two countries almost a decade to pressure the state to adopt legislation that is more in line with the Convention. In Brazil, activists encountered a more favorable political situation after the election of President Lula in 2002, and considerable progress has been made since then. In Uruguay, activists successfully used the Convention to pressure for a legal provision on domestic violence in 1995. Yet, since then progress has been slow and difficult, and Uruguay is still far removed from full compliance. Table 1 gives an overview of national legislation after the adoption of the OAS Convention in 1994. Next, I will go deeper into the national dynamics of setting norms to end violence and how these implied a narrowing down and extending the OAS Convention.
Table 1. Legislation on violence against women in Brazil and Uruguay
After taking office in January 1995, the Social Democratic Cardoso administration ratified the OAS Convention in November 1995, but no efforts were made to translate the Convention to a comprehensive national policy response. Instead, the period after the ratification of the OAS treaty marked erosion of previous gains rather than advancement. The special women's police units created through feminist efforts in the 1980s lost their place and functionality due to a law (9.099, enacted in 1995) that promoted the creation of special crime courts (Adelman and Correa de Azevedo 2012). While this legislation was not directly or explicitly related to the issue of violence against women, it had a major impact on how cases of domestic violence were treated by the state. The courts became the major site for dealing with the complaints brought to the women's police stations. Of all cases handled by the Special Courts, 70% involved domestic violence, of which 50% ended with no criminal charges being pressed (Bandeira 2009, 417). While the special police station was created to address the problem of violence against women and treated individual women as rights-bearing subjects, the special courts instead focused on the family. Courts actively encouraged couples' reconciliation, pushing the problem back into the private sphere. According to Adelman and Correa de Azevedo (2012, 69), prosecutors and lawyers at the special courts often promoted conservative family values and expressed “judgmental attitudes that blame the women themselves for ‘being complicit’ or not knowing how to exercise their rights.” The special courts thus promoted norms that reprivatized domestic violence and put emphasis on the protection of the family. Moreover, the low percentage of prosecution promoted a culture of impunity (Macaulay 2005). This law thus created a norm dynamics that counteracted the OAS norm and led to an erosion of previous gains.
The situation of impunity and the unresponsiveness of the government made the Brazilian lawyers connected to the Latin American and Caribbean Committee for the Defense of Women's Rights (CLADEM) decide to turn to international litigation. Together with the regional human rights organization, the Center for Justice and International Law (CEJIL), they filed a complaint on behalf of Maria da Penha Maia Fernandes. Maria da Penha survived two attempted murders by her husband and, as a result of this aggression, has been paraplegic since 1983. For nearly two decades the case she filed languished in court while Maria's husband remained free. Her petition maintained that the state condoned the situation for more than fifteen years and that it failed to take appropriate measures to prosecute and punish her aggressor despite repeated complaints. In 2001, the IACHR concluded that the Brazilian state had “violated the rights of Mrs. Maria da Penha Maia Fernandes to a fair trial and judicial protection” (IACHR, Report no. 54/01, case 12.051 Maria da Penha Maia Fernandes, Brazil, April 16, 2001). The IACHR asked for a full investigation into her case, compensation, and “the adoption of measures at the national level to eliminate tolerance by the State of domestic violence against women” (IACHR, Report no. 54/01, case 12.051 Maria da Penha Maia Fernandes, Brazil, April 16, 2001). Throughout the proceedings the Brazilian government remained silent and did not file an answer (Roure 2009, 90). Feminist organizations in Brazil then used the presidential election campaigns in 2002 to request attention to the Maria da Penha case. The newly elected President Luiz Inácio Lula da Silva established 2004 as Women's Year through federal law and created the special secretariat of public policies for women. This special secretariat, together with women's organizations, feminist legal experts, and a women's caucus within the Chamber of Deputies, developed and promoted the Maria da Penha Law on Domestic and Family Violence, which passed with President Lula's support in 2006. The new law was strongly based on the Inter-American Convention and framed violence against women as a criminal offense. While the law implied a limitation of the OAS Convention to focus only on violence in the domestic realm of intimate relationships, it also extends the definition of the Convention. It extends the scope of family, defining it “as the community formed by individuals that are or consider themselves related, joined by natural ties, by affinity or by express will” and includes same-sex relations. Moreover, violence against women includes “moral” and material damage. The law established mechanisms for the prevention and reparation of domestic violence (articles 8–10), participation of nongovernmental organizations; conducting research and statistical studies; awareness campaigns; training of public officials on issues of gender, race and ethnicity; and implementation of educational programs in school curricula. In addition, the law created centers of multidisciplinary and comprehensive care for women (article 35) and envisioned rehabilitation and gender re-education programs for offenders (article 45). While the Maria da Penha law implied a major victory, women's organizations have been pushing the state to provide more resources to make the law work (Shadow report 2007). The law ended up prohibiting the alternative measures that had been applied to domestic violence cases since 1995. Instead, acts involving physical injury are automatically criminalized.
A third key moment is the new law on femicide that passed in 2015. After Dilma Roussef was elected president in 2011 she made VAW one of her priorities and increased the number of services, shelters, and special police units. In 2013 she launched the federal program Mulher: Viver sem Violência (Woman: Living without Violence) with an investment of more than $118 million between 2013 and 2014. Additionally, she announced a new law on femicide, which passed in March 2015. The new law (13.104/2015) provides a legal definition for femicide—the killing of a woman by a man because of her gender—which is based on the Declaration on Femicide (2008)of the OAS/CIM Committee of Experts.
In Uruguay, feminist efforts to make violence against women a state issue have only been partially successful in the past two decades. Subsequent governments have been reluctant to adopt a comprehensive approach to violence that passes beyond the domestic sphere. In addition, despite the lobby of Uruguayan feminists for policies that challenge the cultural tolerance of gendered violence, legislation does not specifically identify women as the main victims and survivors of this type of violence. Instead, the Uruguayan government adopted a punitive approach to violence that has little to do with women's rights. Instead, women (and children) are defined as vulnerable subjects in need of state protection. Whereas an initial state response adopted an extremely narrow definition of domestic violence, after the millennium a broad coalition of women inside and outside the state managed to broaden this approach.
The RUDVS and Espacio Feminista started to lobby for the ratification of the OAS Convention and national legislation. These attempts were partially successful. In 1995, domestic violence was included in a new bill on citizen security (nr. 16.707). Article 321bis criminalized “violence or threats that are prolonged in time and cause physical harm that occurs within an affective or family relation.” The limited gender-neutral legal definition of domestic violence remained far from the comprehensive and gendered framing of violence of the OAS Convention that the Uruguayan government ratified in February 1996. The RUVDS and Espacio Feminista continued their advocacy work and jointly developed a comprehensive project that consisted of a study of the incidence of domestic violence, services for victims, and police trainings. This proposal was supported by the ministry of the interior but externally financed by the Inter-American Development Bank. One of the central actors in this project, the director of the national crime prevention office, Gabriela Lopéz Machín, urged the Ministry of Interior to develop public policies to prevent violence. The government responded positively and installed an honorary Interministerial Committee to Formulate a National Plan for the Prevention, Detection and Treatment of Domestic Violence. The committee was headed by Graciela Lopéz Machín and consisted of representatives of the ministries of the interior, education, and public health, the head of the women's special police units, the director of the Women's Institute, and three delegates from the RUVDS (RUDVS 2005). The committee presented their extensive proposal for new legislation in 1999, but the proposal was shelved when the government's legislative period ended in February 2000. In March 2000, a broad coalition of women inside and outside the state successfully urged the new government, again led by the Colorado Party, to adopt legislation on violence (Johnson 2002). In July of 2002, the law on domestic violence (17.514) was passed that characterized domestic violence as a liberty crime and defined any acts “illegally limiting the free exercise of or enjoyment of the human rights of another person” as domestic violence. While the new law still framed violence as a gender-neutral problem, it no longer limited the problem to physical violence and extended the OAS definition to include property-related damage. In addition to the law, the government created the honorary National Council to combat domestic violence consisting of both state and nonstate actors. The council presented a comprehensive national plan in 2002. Yet this plan was not supported by a substantial budget (RUDVS 2005). Since then women's organizations and the OAS have urged the Uruguayan government to broaden its policy focus and develop a more comprehensive approach to violence against women. Very few services and programs are available to address violence, and these are mostly run by civil society and are funded externally (MESECVI 2012).
Feminist action to politicize violence beyond the domestic sphere has been only partially successful. Uruguayan feminists argued that the concentration of government efforts only on domestic violence, rather than those perpetrated within the community and by the state, leaves women utterly unprotected in the public sphere. Uruguayan activists staged protest to ask attention for the issue of sexual harassment in the public space. This has resulted in the adoption of a sexual harassment law in 2009 (18561) that included harassment in the workplace and educational institutions. Yet, like in the case of domestic violence, it is not recognized as a form of gender discrimination. Feminist struggles continue to politicize sexual harassment in the public space.
Institutionalizing norms to end violence against women in Latin America has been a long and ongoing process with progress and setbacks. I have argued that prevailing approaches of global norm diffusion processes do not adequately capture these complex and ongoing dynamics. Instead, building on recent insights, I have proposed an alternative approach to norm diffusion as multilevel and multidirectional in which norms are subject to ongoing contestation and change. Such an approach redirects our attention to processes of norm-setting in a multilevel context instead of focusing on the diffusion of norms between separate and hierarchical levels. This provides us with a number of important new insights.
First, we can see that new ideas and models to address violence against women simultaneously emerged in different domestic contexts but that the creation of new transnational spaces through Encuentros and networking allowed feminists in Latin America to develop a new understanding of violence against women. These new spaces helped to integrate domestic knowledge and experiences of institutionalized violence under military regimes, processes of democratization, and state building and connect these to existing and emerging international frameworks on human rights and gender equality. From the ongoing interactions between differently located activists in these newly created transnational spaces new frames emerged and were co-constructed. This helps us to see national, regional, and international contexts as interrelated and mutually constitutive.
Second, it helps us to see international norms neither as clear starting points that then travel to domestic context, nor as fixed end products that are left unchallenged, once established, by an authoritative international organization. While the adoption of the Inter-American Convention on the Prevention, Punishment and Eradication of Violence against Women (1994) marked an important step in the promotion and institutionalization of norm setting to end violence against women, its definition was narrowed down, bent, and extended in national processes. Feminist attempts to end VAW challenged norms about what is public and private and what should be the role of the state and how rules and procedures need to be altered to prevent and punish gender-based violence. This caused resistance from the actors and systems that were targeted. Governments in Brazil and Uruguay initially developed policies that drastically narrowed down and distorted the progressive framing of the Convention. Policy responses were limited to “domestic violence,” and no measures were adopted to address violence in the sphere of the community or violence tolerated or perpetrated by the state or its agents. Also, state responses ignored the gendered aspects of the violence and did not recognize violence against women as a human rights' violation. This resulted in policy approaches that tended to prioritize the conservation and stability of the family over the rights of women. Yet, we have seen that national struggles and contestations also resulted in the extension of the OAS definition. Norms are thus subject to renegotiation and redefinition. Diffusion theory should therefore be revised to pay more attention to the dynamics of contention that are central to diffusion processes.